Does the good drive out the perfect? The popularization of sprint retrospectives, mini-retrospectives, lightening retrospectives, and other techniques has put “Norm-style” three-day retrospectives in a bit of a shadow. How do facilitators manage these lean activities, and how does one balance and mix them with “Norm-style” events?

The group found that both kinds of retrospectives have value. It is noteworthy that:

• Clients often run their own sprint retrospectives without benefit of tools, grounding, prior groundwork and retrospective tools, so their benefits may be limited.
• The same issues often recur at sprint retrospectives in the absence of a “Norm-style” retrospective, suggesting that some issues go to a level of trust that cannot easily be accessed in the more spontaneous events.
• Visiting a concern at a sprint retrospective has been used as an excuse to skip it at a later retrospective, suggesting that topics that need in-depth treatment and trust considerations may too easily be skipped
• It is likely that few facilitators have the skills to properly conduct a protracted “Norm-style” retrospective.

Attendees summarized their reflections with the following words for further inspiration:

• Retrospective variations
• Possibilities
• Skills
• Prenticeable
• See one, do one, teach one
• Discipline

Willem, Rachel, Jens, Laurent, Angela, Gertrud, Cope

Laurent: Do you know the concept of the canary in the coal mine? I am disturbed by Alistair's interventions on the discussions on retrospectives, where he said we only need half a day. How much time do you need, what are the forces behind Norm's three (my two) days (which itself is a hard sell)? If you're going to do 2-day events, how long of a period should lapse between them? Do you need longer retrospectives for two-year projects? I also wanted to relate a recent reto facilitation experience: 2-day retrospective with a team that had been holding iteration retrospectives. The thing that struck me is that the time line came out rather sparse: there seemed to be a “we've done that before” effect and an inhibition at the points of the process where you build the inventory of events. The team saw that we keep bringing up the same post-its in retrospectives over and over, week after week, and keep solving. Rachel, have you done both?

Rachel: As a facilitator I'm brought in for the longer ones (1-2 days); as a coach I teach them the shorter ones. They are self-taught at doing heartbeat and sprint retrospectives. I had recently I went to [A company] where I was asked to do a retrospective with a team who had been doing sprint retrospectives, and “we don't seem to be solving problems”. “Can we have an interesting exciting format?” No: let's keep it simple, because as a group, you're new to this. So we used time lines and other basic techniques. And they said: what? This isn't what we usually do! I worry that people are dong sprint retrospectives without having the skills to do them. One of their big puzzles is a Big Problem: continuous integration. They are making small steps every time but why they are beating themselves up is that they feel they should be able to solve it more quickly. They are hoping for miracle cures, but can't see the six months of small changes necessary to get there.

That they bring an outsider says something for them.

Surveys often say that people do retrospectives. Maybe they've learned by being in one.

Laurent: How many teams had both?

Rachel: 50 teams had both iterative and big retrospectives. The annual big retrospectives are one day or one and a half days. I've never been able to persuade people to do a three day.

Jens: How often do you hold these, and for how long?

Rachel: 1 hour, or 90 minutes, every two weeks. Angela, what was your experience?

Angela: I was surprised that people were astonished that I had done any preparation. They are trying to do it without understanding what's behind it, or the tools or techniques.

Cope: Has agile brought us to a place where we have become complacent and are using lazy techniques, too easily forgetting the hard work and discipline? We see this all too often when people adopt Agile techniques.

Angela: Well, maybe, but they were succeeding in affecting some change in their organization as a result of sprint reviews.

Laurent: The pattern here is that there is nothing more evil than to adopt fads: Retrospectives are a fad; they don't really know why they are doing them. I would formulate a rule: When you hear “we do X”, where X is one of the nifty toys we play with in this community, ask: “How much work did you put into the skills?”

Cope: There are three kinds of learning: single, douple, and triple loop. Single loop is cause-and-effect; double-loop is at the level of structure and learning how to learn; triple loop is at the level of identity and reaches the level of principles. Do sprint retrospectives stop at single loop learning, and we can get to double loop only with “Norm-style” retrospectives? Do facilitators themselves understand the learning dynamics of the retrospective format beyond the fad level?

Laurent: Maybe we rely too much on gimmicky kinds of things. It seems to be a subtext of whatever we sell is that: It's easy to do, go and do it. That's what's cheapening it. It's not Agile per se; it's that anyone can start doing it. Of course, that's true; but there is another subtext: it takes discipline. Introspection is crucial to success, and if you stop short of that, you're only going through the motions.

Willem: Lots of cultures have local meetings: Swiss town meetings; tribes around the trees in “primitive” cultures, etc. So this might indeed be a cultural meme, contrary to Cope's discounting. There is one culture that does this by sitting down for a long time and not being allowed to move until you've solved your problems; this often has repercussions when nature calls . The moral equivalent of open space is common in native American tribes. The thing I've been doing, is that I train (experientially) to learn about retrospectives *and* I want to improve my course. So I want to hold a retrospective at the end of each day as an example, and so I get feedback on my training. I've been doing that for the past couple of years. It takes some time to pick up the trends. These last only ten to fifteen minutes, and it's been very effective. And maybe for teams as well. Maybe this would be a good time to gather metrics on people who went to the first courses.

Jens I do a similar thing: a retrospective on day 1 of the CSM class, asking how we can improve on day 2. I use a time line and an emotional line.

Willem: First day I explain the temperature reading. First day; complaints/recommendations/puzzles/aspirations; time line; preparations would be (if I remember) I put up the flip chart at the beginning of the day. Some of these are longstanding teams; but most of these are short-term engagements for assorted people in the southern Netherlands. We explain our way of experiential teaching. We also say that we are not the experts.

Cope: Have you ever thought of your course as being a retrospective in its own right?

Willem: Hmmm, that's intriguing… Well, we get green cards for the story-telling: stories. Maybe this course is a retrospective… actually, it's 6. There is something that changes all the time.

Laurent: It's useful to look the longevity of the community that forms. The trainer here knows it will be a two-day course, and knows he may not continue the community, and that's that. For something that's bounded in time yo want a halfway point rule: when we've done 50%, that's the last reasonable time to ask people if they want to change course. Frame things at the beginning: you are in charge of your learning, spend 5 minutes writing your goals for the training. At the end of the day, ask them: How are you doing with respect to your goals? It's important to get people to write things down because it anchors things. I used to do this irregardless.

Angela does something similar: a daily learning stand-up (what did I learn today, what do I want to learn tomorrow; what are my obstacles). They get to see stand-ups. I use this in my 5-day Agile courses at Oxford. By Thursday they learn there are some useful things and structures. It takes a bit longer than 15 minutes; Most people talk, but they are given the option to pass. On the Friday I bring in someone to run a retrospective on the course. I was nervous because, of course, I grade them. Jens: when do you have this? Rachel: at the end of the day. Never has done one in the pub for that course.

Rachel and Angela have both done these in a pub.

Laurent: I run a cycle of training on organizational learning. One of the topics we had was: Feeling free to ask for things. They had it in a bistro. there was a very good occasion for a demonstration; we ordered a round and he asked for a 10% discount, and was told that he could if there are more of you.

Jens: Sometimes you see bonding in two-day courses.

Angela: Sometimes you need to be silly, learn from your mistakes,

Willem: We also do this, and we put ourselves in a position of vulnerability and make it clear to the students that we're not the experts.

Cope: Wow! Maybe vulnerability is a key concept here?

Angela: If you come in as an external facilitator, then I will meet with people to develop vulnerability one-on-one, and talk to them about how that might work in a group setting. How do we do that in a project setting? We do things that allow people to be vulnerable in a retrospective.

Laurent: Half of the retrospective is the prework.

Jim: We did a retrospective at a client where we indeed found that it took several months of interviewing, of meeting with management, and of planning, to be ready for an effective retrospective that would be productive and which would be safe for the participants. We also traveled to another country to interview their clients. We couldn't have done a successful event without these preparations.

Angela; In training courses I use silly games and exercises. In long-term engagements I go deeper and build trust with more preparation.

Laurent: Rhythm and tempo also comes in there. The retrospective is two days, but the engagement is 6 weeks (by the book! by Norm's book!) So I feel uneasy about people who do the quick things. People just don't understand the need for an invitation letter, letters for the interviews, and many other preparatory steps. It can take many weeks. Then you have a post-retrospective meeting or phone call to check on whether the lessons have been taken to heart. That all can take up to six weeks or two months.

Angela: I would do a quick check with everyone on the team, but won't say that I talked to everybody.

Cope: How well-prepared are the trainers? Have people who run sprint retrospectives read the corresponding books for that technique but not Norm's book?

Angela: I think most facilitators haven't read either of them.

Laurent: The way I got into retrospectives was to join was as a back-seat facilitator. Nynke called me one day and said “We're going to facilitate a retrospective for the Dutch XP group.” O.K… So I was able to join, even though I was a novice, a back-seat facilitator. She offered that as the most natural thing in the world to pick a random person in the world as a partner to go in and do this.

Willem: It was the best way in the world to learn about retrospective. My memories differ from Nynke's in terms of the roots of the ideas. My understanding is that Nynke found my wiki, where I had mentioned retrospectives and the possibilities therein, and that led Nynke to believe that there were retrospectives in the Netherlands and so she decided to go with it